The Nature of the Fluorine Exchange Processes in ... - ACS Publications

The length of the Kel-F or Teflon insert was ... 18 0 SEC". 21 SEC-. 300. 1 C SEC-. Figure 1. Experimental and calculated IH ..... SOC., 96, 5385 (197...
0 downloads 0 Views 542KB Size
2161

The Nature of the Fluorine Exchange Processes in Some Alkyl- and Dimethylaminofluorophosphoranes C. G. Moreland,*' G. 0. Doak,*' L. B. Littlefield,' N. S. Walker,2 J. W. Gilje,3 R. W. Braun,2 and A. H. Cowley*2 Contribution from the Departments of Chemistry, North Carolina State Uniaersity, Raleigh, North Carolina 27607, and University of Texas at Austin, Austin, Texas 78712. Receiaed March 19, I975 Abstract: The fluorine exchange processes in (CH3)2PF3 ( l ) , (CH3)3PF2 (2), (CzHs)zPF3 (3),and [(CH3)2N]2PF3 (4) have been investigated by means of ' H , I9F, and 31Pdynamic N M R ( D N M R ) spectroscopy. When the experiments were conducted in Teflon or Kel-F cells, the exchange was found to be intramolecular. The earlier claim (T. A. Furtsch, D. s. Dierdorf, and A . H . Cowley, J . Am. Chem. Soc., 92, 5759 (1970)) that 1 and 2 undergo exchange via an intermolecular (bimolecular) process is therefore not valid. Regardless of the rigor of the purification of 1 and 2, nonreproducible N M R line shapes are observed when the experiments are conducted in Pyrex tubes. Using the technique of matching the observed and computer simulated spectra, it was possible to calculate the rates of, and activation parameters for, the fluorine exchange processes in 1 and 4. T h e fluorine exchange in 1 and 4 proceeds by either the a e or aae process, corresponding to either the M2 or M4 rearrangement mode, respectively (or combinations thereof). Finally, attention is drawn to the fact that, in phosphoranes of the type R3PF2 (e.g., 2). it is impossible to decide on D N M R evidence whether the structures are stereochemically rigid or whether they are interconverting between degenerate conformations.

The literature is replete with examples of N M R detectable ligand exchange processes of pentacoordinate phosphorus compound^.^ In the majority of these examples, ligand exchange has been demonstrated or assumed to be intramolecular in character, the so-called Berry5 "pseudo rotation process" being the mechanism of popular choice. Other studies have, however, revealed the following additional possibilities: (1) intramolecular ligand exchange via "turnstile rotation",6 square pyramidal intermediates,' and other8 mechanisms; (2) intermolecular fluorine e x ~ h a n g e ; ~ (3) solvent assisted exchange;I0 (4) exchange via dimeric The intermediates;] I ( 5 ) impurity catalyzed exchange.'ob%'2 suggestion of intermolecular exchange in (CH3)2PF3 (1) and (CH3)3PF2 (2) was based on (a) the observation that F-P-C-H coupling is absent in the ambient temperature I H spectra, (b) the (bimolecular) substrate concentration dependence, (c) the independence of the results of the method of preparation of the fluorophosphorane, and (d) the insensitivity of the results to the presence of N a F in the N M R tube. However, since that time Moreland, Doak, and LittlefieldI3 have demonstrated that the ligand exchange in the closely related phosphorane ( C ~ H S ) ~ isP intramolecuF~ lar, thereby raising a modicum of doubt concerning the validity of the results which were obtained with the methyl analog. Furthermore, these authors demonstrated that reproducible results cannot be obtained if the N M R experiments are conducted in Pyrex tubes. In view of the foregoing, it seemed necessary to repeat the dynamic N M R ( D N M R ) experiments on 1 and 2 in Teflon or Kel-F N M R cells. The primary purpose of the present paper is to point out that the fluorine exchange in 1 and 2 is intramolecular and not intermolecular as claimed p r e v i o ~ s l y The . ~ ~ earlier results on these compounds are artifactual and presumably due to the reaction of 1 and 2 with the Pyrex glass, the fluorine exchange being catalyzed by the consequent adventitious impurities. It has also been possible to calculate the activation parameters for the intramolecular fluorine exchange in l and [(CH3)2N]2PF3 (4) by standard line-shape analyses of the ' H and 3 ' P D N M R spectra. Finally, the intramolecular fluorine exchange in 1, 4, and (CzH5)2PF3 (3) and the temperature insensitivity of the N M R spectra of 2 (and its analogs) are discussed on the basis of rearrangement modes14 and observable processes.Is

Experimental Section The dialkyltrifluorophosphoranes 1 and 3 were prepared by the action of SbF3 on R2PC1I6 or R4PzS2" (R = CH3 or C ~ H S )and , 2 was prepared by the reaction of (CH3),P with SF4 at -78 OC.'* Compound 4 was prepared by the action of (CH,)zNSi(CH3)3 on PF5.I9 All compounds were purified by fractional condensation in a Kel-F/stainless steel vacuum line until they were tensimetrically homogeneous. The ir spectral frequencies were in good agreement with the literature values.20 All samples were stored over anhydrous N a F in a Kel-F/stainless steel vessel prior to use. The N M R samples were prepared by condensing the appropriate fluorophosphorane into a piece of 3 mm 0.d. flexible Kel-F or Teflon tubing a t - 196 "C which had been sealed a t one end. The Kel-F or Teflon tubing was attached to the vacuum system via a stainless steel valve, and the vacuum seal was effected by bringing a hot coil of nichrome wire near the Kel-F or Teflon tubing. After this preliminary sealing procedure, a final seal was effected by squeezing the top of the sample tube with a pair of warm tongs. During the sealing operations, the sample was maintained at -196 OC. After allowing the sample tube containing the fluorophosphorane to warm up to ambient temperature, the sealed flexible tube was inserted into a standard 5-mm N M R tube containing a small amount of the external standard. The length of the Kel-F or Teflon insert was such that it was possible to place a cap on the external N M R tube. In order to obtain reproducible line shapes, it is important to avoid any contact of 1 or 2 with Pyrex glass during either the purification or sample preparation stages. Furthermore, it is essential that the Kel-F or Teflon cells containing 1 or 2 be removed from the Pyrex N M R tubes when N M R experiments a r e not being conducted. Presumably it is possible for 1 or 2 to diffuse outside of the Kel-F or Teflon cell and to come into contact with the Pyrex tube. Similar precautions were taken with 3 and 4. N M R Spectra. All spectra were determined on a Varian Associates HA- 100 spectrometer equipped with variable-temperature and heteronuclear decoupling accessories. Probe temperatures were calibrated using a Wilmad N M R thermometer. The spectrometer frequencies for the ' H , I9F, and 3 ' P spectra were 100, 94.1, and 40.5 MHz, respectively. Line-Shape Calculations. Theoretical spectra were calculated using the many-site program N M R L S which is based on the equations of Anderson and Kubo.21 The program was devised by Professor M . SaundersZ2and modified for a C D C 6400/6600 computer. Input included the frequencies of the lines measured from an arbitrary zero, their natural line widths and relative intensities in the absence of exchange, and a transition probability matrix.

Results A selection of ' H , I9F, and 3 1 P N M R spectra for 1, 3, and 4 may be found in Figures 1-5. A summary of the

Moreland, Doak, Cowlej!,et al.

/

Fluorine Exchange Processes

2162 EXPERIMENTAL

CALCULATED

8 5"

900 SEC.'

75"

52 5 SEC-'

2 65O

B.

J

i . i

18 0 SEC"

Figure 3. I9F N M R spectra of neat (CH3)2PF3 (1) in a Teflon cell with (a) ' H decoupler on and (b) IH decoupler off.

21 SEC-

300

1 C SEC-

Figure 1. Experimental and calculated IH D N M R spectra of (CH3)2PF3 ( 1 ) . The experimental data were obtained for neat 1 in a Kel-F cell.

30" Figure 4. 'H WMR spectra of neat ( C ~ H S ) ~ (3) P Fin~ a Kel-F cell at various temperatures. Only the CH2 resonance is shown.

date the experimental data was one in which the equatorial fluorine of 1 or 4 had an equal chance of intramolecular exchange with an axial fluorine.23 For the ' H spectra of 1, the transition probability matrix took the form where the line Line Numb e r l Figure 2. Experimental "P D N M R spectra of neat (CH?)?PF?(1) i n a Teflon cell. The 82-, 91-. and 100 OC spectra were run with the ' H decoupler on. The 32 OC spectrum was run with the IH decoupler off.

N M R data for 1-4 is presented in Table I. The simulated spectra which are displayed beside the experimental ' H spectra in Figures 1 and 5 were computed for the indicated rate constants by means of the N M R L S program (see Experimental Section). For the ' H (and 3 ' P ) spectra, the only probability matrix that would accommoJournal of the American Chemical Society

/

98.8

/

2

3

1 1 0 0 2 0 0 1 3 0 0 . 5 0 . 5 4 0 0 0 5 0 0 0 6 0 0 0 7 0 0 0 8 0 0 0 9 0 0 0 1 0 0 0 0 1 1 0 0 0 1 2 0 0 0

April 14, 1976

4

5

0 0

0 0

6

0 0 0 0 0 0 . 5 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 . 5 0 0 0 0 0 0 0 0 0

7

8

9

0 0 . 0 0 0 0 0 0 0 0 . 5 0 0

0 0 0 0 0 0 0 0

0 0 0 1 0 0 1 0 0 0 . 5 0 0 0 0 0 0

10

1112

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 . 5 0 . 5 0 1 0 0 0 0 1

2163 and 8; however, as pointed out above, these are overlapping with lines 4 and 9, respectively). Using the k values from Figure 1, the best straight-line Arrhenius plots afforded the following activation parameters for 1: E , = 15.5 kcal/mol, Act333 = 17.8 kcal/mol, A S + = -8.8 eu from the 'Hspectra. The values E , = 15.1 kcal/mol, ACT359 = 17.3 kcal/ mol, A S ? = -8.1 eu were obtained from the 3 ' P spectra. The use of a similar approach for 4 yielded the activation parameters, E , = 20.2 kcal/mol, Act343 = 19.6 kcal/mol, AS? = -0.1 eu from the ' H D N M R spectra (Figure 5).

C A LC U LATE0

EXPERIMENTAL

/ I JI

---i:I 70°

2.5 SEC-I

J L 1.0 SEC-'

60°

/

50° Figure 5. Experimental and calculated ' H DNMR spectra of [(CH3)2N]2PF3 (4). The experimental data were obtained for neat 4 in a Kel-F cell.

numbers correspond to the ground state 'H spectrum (Figure l ) , reading from left to right. Note that lines 4 and 5, and 8 and 9 occur at the same frequency. The matrix that was employed for the D N M R data of 1 has been published e l ~ e w h e r e . 'In ~ , the ~ ~ case of 1, a key feature of this rearrangement process is the fact that, as predicted, lines 1 and 12 of the ' H spectra remain sharp (as should lines 5

Discussion The salient feature that emerges from the present study is the observation that, when the N M R experiments for 1 and 2 are done in Kel-F or Teflon cells, there is no evidence f o r an intermolecular exchange process.2s The earlier experiments were conducted in Pyrex glass N M R tubes and the erroneous conclusion that 1 and 2 exhibit intermolecular exchange was based inter alia on the absence of 'H-CP-I9F coupling in the ' H N M R spectra. Furthermore, ,we find that it is absolutely essential to avoid Pyrex glass in both the purification and sample preparation stages. Regardless of the rigor with which the samples of 1 and 2 a r e purified, variable and nonreproducible results are obtained if the substrates are placed in Pyrex glass N M R tubes. In Kel-F or Teflon tubes, both 1 and 2 retain 'H-C-P-I9F coupling up to 85 OC (Figure 1).26 The intramolecular nature of the exchange process in 1 receives additional confirmation from the following facts: (a) only an intramolecular exchange transition probability matrix will accommodate the observed ' H and 31PD N M R spectral changes, and (b) the average 'H-C-P-19F and 31P-19Fhigh-temperature coupling constants correspond to the appropriately weighted averages of the low-temperature axial and equatorial values. The retention of 'H-C-N-P-I9F coupling in 4 indicates that the fluorine positional interchange process is also intramolecular in this molecule, a deduction which is confirmed by the fact that the ' H D N M R spectra (Figure 5) can be simulated only with an intramolecular exchange probability matrix. I n the case of the ethyl derivative 3, the ' H D N M R spectra (Figure 4) were somewhat too complex for lineshape analysis. Furthermore, no attempt has been made to evaluate all the coupling constants. However, it is apparent from the 'H D N M R spectra of the methylene protons that 'H-C-P-19F coupling is preserved up to 70 OC, thereby indicating that the fluorine exchange in 3 is also intramolecular. The activation parameters for the intramolecular fluorine exchange processes in 1 and 4 are comparable to those which have been observed for other trifluorophosphoranes

Table 1. Summary of NMR Data for Trifluorophosphoranes Temp, "C

Conipd (CH J 2 P F 3

(1)

30

Chemical shiftsa 'H

8.525

Coupling constants, Hzb

31P

"F

104

4.14 (a)

JP F

JPCH

JHCPF

17.0

JHCPF, = 13.0

17.2 17.5

JHCPFe = 3.3 JHCPFC= 9.5 11.4

JPNCH

JHNCPF

J P F=~777

86.2 (e) W,),PF, (CHJ,PF, C,H, ),PF 3

(1)

(2) (3)

85 30 30

[(CH,),N],PF,

(4)

30

8.525 8.545 8.92 (CH,) 8.05 (CH,) 7.405

JPF,=966

10.8 J H N C P F ~= 2.6

JHNCPFe = 1.6 uproton chemical shifts in T units relative to external Me,%; ,'P chemical shifts in parts per million upfield from external P,O,; 19F chemical shifts in parts per million upfield from external CFCl,. ba = axial; e = equatorial. CDenotes average value.

Moreland, Doak, Cowley, et al.

/ Fluorine Exchange Processes

2164 Table 111. Rearrangement Modes, M f , and Observable Processes, OPib, for Various Phosphoranes

Table 11. Intramolecular Fluorine Exchange Barriers in Tritluorophosphoranes

A c t , kcal/mol

Compd (CH,),PF, [(CH,),Nl ,PF, (C,H,),PF, H2PF

3

CF3(H)PF Cl,PF, Br,PF, (CF,),PF,

(1)

(4)

17.8 19.6 18.7 10.2 6.3 E , 7 . 2 i 0.5 E , = 7.2 i 0.5 Fluxional at - 1 2 0 ° C

333 343 379 218 133

a

Compd

a

b c

PA,B(EQ)~ PA,B,(EQ)d PA,B,(EQ)~ PA,B(AXId

c d d

M C M, L O P ,

1 1 I 1

1 0 0 0

M,

M,

M,

(Table 11). For X2PF3 molecules, the barriers depend on the substituent X in the order (CH3)2N > C6H5 > CH3 > H > CI Br > CF3. As with the XPF4 the factors which are responsible for the magnitudes of the barriers are not immediately apparent. (Note, however, that the orders are different for the XPF4 and X2PF3 molecules.) r-Bonding has been suggested2' to be an important contributor to the barrier magnitude in XPF4 compounds such as (CH3)2NPF4 and CH3SPF4. It is interesting to note that [(CH3)2N]2PF3 also possesses a high barrier to fluorine exchange. However, the r-bonding postulate cannot be employed to explain the position of H2PF3 in the above order. Attention is drawn next to a discussion of the fluorine positional exchange processes in 1 and 4. A t the outset it might be stressed that the rearrangements of pentacoordinated molecules can proceed via an almost infinite variety of mechanisms including intramolecular, intermolecular, collision activated exchange, solvent assisted exchange, and others. However, since D N M R data are interpreted within the framework of the "jump model" (i.e., transitions are effectively instantaneous),28 they cannot provide mechanistic information; rather such data are best discussed in terms of rearrangement modes (Mi),14which are, of course, merely permutational in character. Furthermore, due to the D N M R spectroscopic indistinguishability of several of the rearrangement modes, it is convenient to effect a further classification into observable processes (OPi).i5 The modes and observable processes for XPF4, X2PF3. and X3PF2 molecules are summarized in Table Ill. As an example, consider first the rearrangements in the PA4B(EQ)29 molecule, (CH3)2NPF4. The D N M R data7a.30can be accommodated by either the aeae3I (mode M I ) or aexae (mode M5) intramolecular process. The experimental data provide no means of distinguishing between these two rearrangements; hence modes M I and M5 are included in the same observable process (OP,). In the case of the PA3B2(EQ)29 molecules 1 and 4, the only probability matrix (vide supra) which is capable of reproducing the experimental ' H and 3 i P D N M R spectra is one in which the equatorial fluorine ligand23 has an equal probability of exchanging with either of the axial fluorine ligands, i.e., the ae3I or aae rearrangement. In turn the ae

4 2 0 3

1 1 1 1

4 2 0 3

R = CH,, or (CHJ,N

and aae rearrangements correspond to modes M2 and M4, respectively. Furthermore, since these modes are D N M R indistinguishable, they are included in the same observable process (OP2). Previously it has been shown the H2PF3 and

/ 98:8 / April

1 0

o

0

12 6 2 8

3 2 1 2

-K

M, aeaee

-

M, Total Total modes OP's

e

upresent work. b Reference 1 3 and unpublished observations. CReference 32. d W . Mahler and E. L. Muetterties, fnorg. Chem., 4, 1520 (1965). =Reference 4a.

Journal of the American Chemical Society

I-:Kl

Temp, K Ref

M, aae

M2

M3

ae

aa

M, aexae

u s e e ref 14 for a discussion of rearrangement modes. bSee ref 15 for a discussion of observable processes. CIdentity operation. dSee ref 29 for definition of symbols. eSee ref 31 for definition of symbols.

CF3(H)PF3 also rearrange by the same observable process.32 Finally, attention is drawn to the PA2B3(EQ)29 molecule, 2. Reference to Table I11 indicates that there is only one observable process (OPo) if it is assumed that the placement of a B ligand in an axial position would require a prohibitively large amount of energy. Unless intermolecular or impurity catalyzed exchange is extant, this implies that the N M R spectra of molecules of the type R3PF2 cannot exhibit temperature sensitivity since the observable process OPo comprises only the identity operation and the aa3I process (mode M3), both of which yield a product which is identical with the initial structure.

Acknowledgments. The authors are grateful to the N a tional Science Foundation (Grant G P 38027X) and the Robert A. Welch Foundation for generous financial support. Appreciation is also expressed to Mr. Billy Roberts for obtaining the 3 1 PN M R spectra. References and Notes (1) North Carolina State University. (2) The University of Texas at Austin. (3) On sabbatical leave from the Chemistry Department, University of Hawaii, Honolulu, Hawaii, during the 1972-1973 academic year. (4) See, e.g., (a) E. L. Muetterties, W. Mahler, and R. Schmutzler, lnorg. Chem., 2, 613 (1963); (b) R. Schmutzier, Adv. fluorine Chem., 5, 31 (1965); (c) R. Schmutzler, Halogen Chem., 3 1 f f(1967); (d) K. Mislow, Acc. Chem. Res., 3, 321 (1970); (e) E. L. Muetterties. ibid., 3, 266 (1970); (f) A. T. Balaban, Rev. Roum. Chim., 18, 855 (1973); (9) M. Gielen, "Stereochimie Dynamique", Scientific Publications Division, Freund Publishing House, Ltd., Tel Aviv, Israel, 1973; (h) W. G. Klemperer in "DNMR Spectroscopy", F. A. Cotton and L. M. Jackman. Ed., Academic Press, New York, N.Y., 1975; (i) R. Luckenbach, "Dynamic Stereochemistry of Pentacoordinated Phosphorus and Related Elements", Georg Thieme Veriag, Stuttgart, 1973. ( 5 ) R. S. Berry, J. Chem. Phys., 32, 933 (1960). (6) (a) I. Ugi, D. Marquarding. H. Klusacek, G. Gokel and P. Gillespie, Angew. Chem., Int. Ed. Engl., 9, 703 (1970); (b) I. Ugi, D. Marquarding. H. Klusacek, P. Gillespie, and F. Ramirez. Acc. Chem. Res., 4, 288 (1971); (c) F. Ramierz and I. Ugi. Adv. Phys. Org. Chem., 9, 25 (1971). (7) (a) M. Eisenhut, H. L. Mitchell, D. D. Traficante, R. J. Kaufman. J. M. Deutch. and G. M. Whitesides, J. Am. Chem. SOC.,96, 5385 (1974); (b) G. M. Whitesides, M. Eisenhut, and W. M. Bunting, ibid., 96, 5398 (1974). (8) R. R. Holmes, Acc. Chem. Res., 5, 296 (1972). (9) (a) T.A. Furtsch, D. S. Dierdorf, and A. H. Cowley, J. Am. Chem. SOC.. 92, 5759 (1970); (b) H. Dreeskamp and K. Hildenbrand, Z.Naturforsch. 8, 26, 269 (1971). (10) (a) J. D. Macomber, J. Magn. Reson., 1, 677 (1969); (b) J. A. Gibson, D. G. ibbott, and A. F. Janzen, Can. J, Chem., 51, 3203 (1973). (11) J. I. Musher, Tetrahedron Lett., 1093 (1973).

14, 1976

2165 (12)For a discussion of impurity catalyzed exchange in analogous sulfur compounds, see W. 0. Klemperer, J. K. Krieger, M. D. McCreary, E. L. Muetterties. D.D. Traficante, and G. M. Whitesides, J. Am. Chem. SOC.. 97, 7023 (1975). (13)C. G. Moreland. G. 0. Doak, and L. B. Littlefield, J. Am. Chem. SOC., 95, 255 (1973). (14)J. I. Musher. J. Am. Chem. SOC.,94, 5662 (1972). (15)(a) J. I. Musher. J. Chem. Educ., 51, 94 (1974);(b) J. I. Musher and W. C. Agosta, J. Am. Chem. SOC.. 96,1320 (1974). (16)R. Schmutzler. lnorg. Chem., 3,421 (1964). (17)R. Schmutzler, Inorg. Chem., 3,410 (1964). (18)The procedure employed was similar to that described by W. Mahler, lnorg. Chem.. 2, 230 (1963). (19)R. Schmutzler, J. Chem. SOC., Dalton Trans., 2687 (1973). (20)(a) A. J. Downs and R. Schmutzler, Spectrochim. Acta, Part A, 23, 681 (1967);(b) J. Grosse and R. Schmutzler. Phosphorus, 4, 49 (1974). (21)(a) P. W. Anderson, J. Phys. SOC.Jpn., 9, 316 (1954);(b) R. Kubo, hid., 9,935 (1954);(c) Nuovo Cimento, Suppl., 6, 1063 (1957). (22)(a) M. Saunders. Tetrahedron Left, 1699 (1963):(b) M. Saunders in "Magnetic Resonance in Biological Systems", A. Ehrenberg, B. G. Malmstrom, and T. Vanngard, Ed., Pergamon Press, Oxford, 1967,p 85. (23)Electron diffraction studies indicate that the methyl groups in 1 adopt the equatorial sites of a trigonal bipyramid [see K. W. Hansen and L. S. Bartell, lnorg. Chem., 4, 1775 (1965)l.A similar structure may be as-

sumed' for 4 on the basis of NMR (ref 19)and photoelectron spectroscopic data [A. H. Cowley, M. J. S. Dewar, D. W. Goodman, and J. R. Schweiger. J. Am. Chem. SOC., 95, 6506 (1973)]. (24)C. S.Johnson and C. 0. Moreland. J. Chem. Educ., 50,477 (1973). (25) A similar conclusion has been arrived at independently by 0. Schlak and R. Schmutzler. The authors are grateful to Professor Schmutzler for informing us of this work. (26)The proton spectra of 2 in the range 30-85 OC are essentially the same as Figure la of ref 9a. (27)E. L. Muetterties, P. Meakin, and R. Hoffmann. J. Am. Chem. SOC.,94,

5674 (1972). (28)For a discussion of the effects of intermediates on DNMR spectra, see ref 7 and references therein. (29)AX or EQ denotes respectively the axial or equatorial location(s) of the B ligand(s) within a trigonal bipyramidal phosphorane structure of general formula PA,,B+", n = 3,4. (30)G. M. Whitesides and H. L. Mitchell, J. Am. Chem. SOC.,91, 5384

(1969). (31)The letters a and e symbolize axial and equatorial ligands, respectively, within a trigonal bipyramidal framework. The various sequences of these letters correspond to the ligand permutations which are depicted in Table 111. (32)J. W. Gilje, R. W. Braun, and A. H. Cowley, J. Chem. SOC.,Chem. Commum, 15 (1974).

Stereochemistry of Eight-Coordinate Mixed-Ligand Complexes of Zirconium. 11. Characterization and the Crystal and Molecular Structure of Nitratotris( acetylacetonato)zirconium(IV)' ,2 E. Gordon Muller,38Victor W. Day,*3band Robert C. Fay*3a Contributionf r o m the Departments of Chemistry, Cornell Uniuersity. Ithaca, New York 14853, and Uniuersity of Nebraska, Lincoln, Nebraska 68508. Received July 7, 1975

Abstract: The crystal and molecular structure of nitratotris(acetylacetonato)zirconium( IV), Zr(acac)3(N03), has been determined by single-crystal x-ray diffraction and has been refined (anisotropically for Zr, N , 0, and C atoms: isotropically for H atoms) by full-matrix least-squares techniques to R I = 0.030 and R2 = 0.032 using 51 10 independent diffractometer-recorded reflections having 2 8 M & , < 71° and I > 3 4 1 ) . The compound crystallizes in the centrosymmetric monoclinic space group P21/c with four molecules in a unit cell of dimensions: a = 9.204 ( I ) , b = 15.648 (2). c = 13.471 ( I ) A, p = 91.364 (8)' (pcalcd = 1.543, p&sd = 1.538 ( 5 ) g/cm3). The crystal contains discrete eight-coordinate molecules in which the bidentate nitrate ligand spans an _a edge and the three bidentate acetylacetonate ligands span b, m, and g edges, respectively, of a (necessarily distorted) D2d-42m dodecahedron. Distortions a r e in the direction of a pseudo-seven-coordinate pentagonal bipyramid. Complexing bonds to the acetylacetonate ligands are systematically shorter than those to the nitrate ligand, averaging 2.141 ( I ) and 2.366 (2) A, respectively. The averaged length for all eight Z r - 0 bonds is 2.197 A. Differences in the two Z r - 0 bond lengths within a particular chelate ring are propagated in the C-0, C-C, and N - 0 bonds of the ligands. The acac ligand "bite" shows significant variation with the type of polyhedral edge it spans: m (2.618 (2) A) < g (2.689 (2) A ) < b (2.786 (2) A). The ligands are planar, and the acetylacetonate methyl groups adopt a conformation in which one methyl hydrogen atom and the -CH= hydrogen atom are eclipsed. The relative merits of the observed CI-abmg stereoisomer and other possible stereoisomers are discussed in terms of ligand bite, polyhedral edge lengths, and nonbonded contacts. In inert solvents, Zr(acac)3(NO,) is a monomeric nonelectrolyte which is stereochemically nonrigid on the N M R time scale. Retention of coordination number eight in solution is suggested by the similarity of solid-state and solution-state infrared spectra.

This is the second of two papers dealing with the structure and stereochemistry of eight-coordinate mixed-ligand nitrato(acety1acetonato)zirconium complexes of the type Zr(acac)z(NO3)2 and Zr(acac)3(N03). The single-crystal x-ray study reported in Part I ' revealed that Zr(acac)2(N03)2 is an eight-coordinate complex in which the acetylacetonate and nitrate ligands (both bidentate) span the four m edges of a (necessarily distorted) D2d-j2m dodecahedron (mmmm stereoisomer4). Each BAAB trapezoid of the dodecahedronS contains one acetylacetonate and one nitrate ligand; thus the approximate molecular point group symmetry is C2-2. The ligand wrapping pattern observed for Zr(acac)z(NO3)2 and the dimensions of the coordinaMuller, Day, Fay

/

tion polyhedron suggest that the relatively large bite of the acetylacetonate ligand does not permit two acac ligands to be located on the same trapezoid of a ZrOg dodecahedron. In support of this view, we note that Zr(acac)a adopts the alternate eight-coordination polyhedron, a D4d-82m square antiprism, with the ligands located on the s edges (ssss stereoisomer).6 In view of the different stereochemistries exhibited by Zr(acac)2(N03)2 and Zr(acac)4, it was of interest to determine the structure of the intermediate mixed-ligand complex Zr(acac)3(N03). Neither the dodecahedral mmmm stereoisomer nor the antiprismatic ssss stereoisomer is likely for Zr(acac)3(N03); the former requires that two acac li-

Stereochemistry of Eight-Coordinate Mixed- Ligand Complexes of Z r